The primary goal of this project is to extend the established role of FMISO PET in evaluating hypoxia in head and neck cancer and to assess its ability to predict response to therapy. In spite of advances in the diagnosis of head and neck cancer, a vast majority of patients are still diagnosed in advanced stages and show poor response to treatment. The clinical course of these neoplasms is difficult to predict based on current prognostic criteria. Current management modalities involve surgery and radiation therapy in different combinations; with nearly half the patients receiving radiotherapy. Advanced head and neck cancer frequently exhibits hypoxia that imparts resistance, negatively affecting treatment response and outcome. Advances in radiation treatment planning and delivery promise the ability to deliver higher radiation dose to a small volume of tissue at high risk of failure (such as hypoxic tissue) without increasing morbidity. Our experiments are designed to non-invasively quantify regional hypoxia in head and neck cancer by [F-18] fluoromisonidazole (FMISO) PET. The specific aims are: (1a) Correlate FMISO measures of hypoxia with treatment outcomes; and investigate the best image metric that describes tissue hypoxia and clinically useful predictors of disease course and treatment; (1b) Test the change in FMISO uptake as a predictor of treatment outcome; (2a & b) Correlate pre-treatment FMISO uptake with biological markers of hypoxia, angiogenesis, glucose metabolism and cellular proliferation to establish its role as a prognostic variable. FMISO PET images will be correlated with biomarkers and genomic expression on microarray in tissue specimens to create a prognostic profile. Coregistered FMISO PET and CT images will be used to direct sampling of resected specimens. (3) Fusion of FMISO PET/CT images to treatment planning CT using validated coregistration techniques to project hypoxic tissue on anatomic imaging. FMISO PET images will be aligned to our treatment planning systems to evaluate the feasibility of incorporating hypoxic regions in treatment planning and delivering 'boost' radiotherapy to these regions. This information will not be used to alter patient treatment, but will evaluate the synergy between FMISO PET + IMRT, in producing clinically optimal treatment plans. We will also analyze failure patterns in FMISO-avid regions that could have undergone dose escalation. Successful completion of this project will result in a better understanding of the problem of hypoxia in head and neck cancer and the role of FMISO PET as a prognostic variable laying the foundation for large scale studies and more rational treatments based on tumor microenvironment.